Test-retest Reproducibility of Right Heart Parameters of Chamber Size and Function

Right ventricular (RV) size and function play a crucial role in determining patient functional status and prognosis in various conditions. The quantification of RV dimensions and function is part of any routine imaging workup involving echocardiography and cardiac magnetic resonance. Complex clinical decision-making, such as to select candidates for various interventions (repair procedures for treating tricuspid regurgitation, ventricular assist devices, etc) and imaging follow-up to identify the optimal timing for intervention require an accurate and reproducible approach to RV evaluation. However, due to its complex shape, thin and highly trabeculated wall, the RV is more challenging and technically difficult to evaluate in comparison to the left ventricle (LV). In the last years, RV strain by speckle-tracking echocardiography (STE) and ejection fraction by three-dimensional (3D) echocardiography emerged as superior metrics of RV systolic performance, overcoming some of the limitations of conventional echocardiographic parameters. To date, cardiac magnetic resonance (CMR) has been considered the gold standard for the assessment of RV volumes and ejection fraction. More recently, RV strain as a more sensitive and less load-dependent metric of RV performance than ejection fraction can be quantified by CMR feature tracking, yet its clinical value remains to be demonstrated.

The reliability of advanced echocardiographic and CMR parameters at repeated measurements is not fully clarified. For echocardiography, the variability of the measurement is more complex, as it depends both on acquisition and reading variability, but also closer to the real life setting than observer variability. Much more limited data exist on the test-retest reliability of RV, right atrial (RA) and tricuspid annulus (TA) parameters than on their observer variability and than of the equivalent left-sided parameters. Test-retest reliability of measurements is crucial for follow-up studies and for clinical monitoring of patients to detect a significant change in ventricular performance, as well as to assess the outcome of various therapies on tricuspid valve (TV) and RV size and function.

The investigators aim to explore in a multicenter setting the test-retest reproducibility of RV and RA size and function, and of tricuspid annulus (TA) size by echocardiography and CMR.

The primary aim of the study is to compare the test-retest reproducibility and agreement of advanced echocardiographic parameters of RV and RA size and function, and of tricuspid annulus (TA) size against the respective parameters obtained by conventional echocardiography and by CMR (where applicable).

Secondary aims:

1. Compare the test-retest reproducibility and agreement of echocardiographic measurements with corresponding CMR measurements: 3D right ventricular ejection fraction vs CMR right ventricular ejection fraction, 2D TA diameter vs CMR TA diameter etc, in individual centers and in core lab
2. Compare the test-retest reproducibility of RV measurements with corresponding LV measurements (i.e. right vs left ventricular ejection fraction, right vs left ventricular global longitudinal strain etc) using echocardiography and CMR in individual centers and in core lab
3. Compare reproducibility of measurements between 2 different analysis software applied on the same images (at core lab)
4. Assess the impact of observer experience and center volume (n° studies /year) on the measurement reproducibility on semi-automated vs fully-automated analyses; to evaluate the potential role of training on improving reproducibility, the inter-observer reproducibility of tested parameters will be compared between 10 expert readers (averaged measurements) and 10 new users (trainees).

Methodology:

Multicenter cross-sectional prospective study of consecutive patients fulfilling the study enrolment criteria, with clinical indication to perform echocardiography and CMR study, with a relatively fair distribution of right ventricular ejection fraction: normal (25%), mildly reduced (25%), moderately reduced (25%) and severely reduced (25%), will be enrolled in each center.

The time interval between echo and CMR studies should be < 48 h, in the absence of any significant change in clinical status or therapy between the 2 studies.

For echocardiography, 2 sets of acquisitions by 2 different operators will be performed at the beginning and at the end of a routine study (at least 20 min apart). This set will include:

• RV focused 4-chamber view (FR 50-80 fps)
• M-mode tricuspid annulus plane systolic excursion (TAPSE)
• Tissue Doppler S-wave
• LV 4-, 2-chamber and apical long-axis (FR 50-80 fps)
• Conventional 4-chamber dedicated for left and right atrium
• 3D dataset of RV (including TA and right atrium)
• 3D dataset of LV (including left atrium)
• RV systolic pressure by continuous wave Doppler

Test-retest measurements will be done independently by the same two operators that acquired the echo images.

For CMR, 2 sets of acquisitions by 2 different technicians including 4-chamber view cine and two sets of short-axis stacks for RV will be acquired at the beginning and at the end of a routine study (at least 20 min apart) after having removed the patient from the scan and his/her repositioning on the table for the second set of acquisitions.

Test-retest measurements will be done independently by two operators analyzing blindly the two acquisitions.

Measurements obtained in each center will be compared with core lab measurements.

Blood pressure and heart rate will be recorded before and at the end of the study.

Demographic, anthropometric and clinical data will be collected.

Based on non-inferiority of 3D right ventricular ejection fraction reproducibility as compared to CMR right ventricular ejection fraction reproducibility (ICC=0.82), the calculated minimal sample size with 5% margin is 183 patients. Taking into account 15% risk of poor quality images or missing value, the sample size was set to at least 215 patients to be enrolled in 6 months.

A minimum of 20 European centers will be invited to collect echocardiography and CMR data from at least 12 subjects/center with a relatively fair distribution of RV ejection fraction by CMR: normal (25%), mildly reduced (25%), moderately reduced (25%) and severely reduced (25%) to be enrolled in 6 months.